Continuation of a pre-breeding program for improving wheat yield

Slabbert, Johannes Diederick (2020-04)

Thesis (MScAgric)--Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: Wheat is a crop that has been cultivated around the globe for centuries and forms a substantial portion of the population’s diet, particularly in third world countries. Food security is under major stress with the human population increasing, thus it is important to increase the amount of wheat produced to meet the demand. Higher yields along with better quality can be reached by an increased breeding efficiency through research and development of breeding techniques. The aim of this study was to identify increased yield-related traits and introducing them into the marker-assisted recurrent selection (MS-MARS) facilitated pre-breeding program of the Stellenbosch University Plant Breeding Lab (SU-PBL) to breed wheat lines higher yields. Germplasm was identified through literature that could benefit the breeding program by having traits related to increased yield. The standard set of the SU-PBLs molecular markers were used to make informed decisions during the selection process. Remote sensing by a Remote Pilot Aircraft System (RPAS) was used to perform detailed observations of wheat and was compared to traditional instruments. Field- and post-harvest phenotyping was done to aid in the selection of lines with high-yielding traits. Lines with high yield-related traits were identified by using the different techniques and were introduced into the MS-MARS scheme. R-squared values of linear regression models displayed that the RPAS data could not predict the phenotypic data in the field, except for yield. Scatter plot matrices shown that there was no correlation between the data captured by the traditional instruments and the RPAS. The field and post-harvest data indicated that a nearest neighbour analysis (NNA) was the best option during this study as there were field trends and the data was used in the selection process. Future studies should include the addition of molecular markers that correlate with increased yield-related traits to make more informed decisions during selection. The development of cameras and software for remote sensing will definitely benefit the tool. Additional vegetative indices can be explored and the model can be improved over time by the addition of data.

AFRIKAANSE OPSOMMING: Koring is 'n gewas wat al eeue heen wereldwyd verbou word en vorm 'n belangrike deel van basiese voeding, veral in die derde wêreld. Voedselsekerheid is onder groot druk terwyl die bevolking toeneem, en dit is dus belangrik om die hoeveelheid koring wat geproduseer word, te verhoog om aan die vraag te voldoen. Hoër opbrengste en beter gehalte kan bereik word deur 'n verhoging in teelprogramme se effektiwiteit deur navorsing en ontwikkeling van teeltegnieke. Die doel van hierdie studie was om verhoogde opbrengs verwante eienskappe te identifiseer en dit in die merker bemiddelde herhalende seleksie (MS-MARS) gefasiliteerde voorteelprogram van die Universiteit Stellenbosch Planteteeltlaboratorium (SU-PBL) in te sluit ten einde lyne met hoër opbrengs te teel. Kiemplasma is geïdentifiseer deur middel van literatuur wat die teelprogram kan bevoordeel deur eienskappe te hê wat verband hou met verhoogde opbrengs. Die standaardstel van die SU-PBL se molekulêre merkers is gebruik om ingeligte besluite te neem tydens die seleksie proses. Afstandswaarneming deur 'n ‘Remote Pilot Aircraft System’ (RPAS) is gebruik om gedetailleerde waarnemings van koring uit te voer en dit is vergelyk met tradisionele instrumente. Fenotipering van veld- en na-oes eienskappe is gedoen om te help met die seleksie van lyne met eienskappe met 'n hoë opbrengs. Lyne met hoë opbrengsverwante eienskappe is geïdentifiseer met behulp van die verskillende tegnieke en is in die MS-MARS-skema gebruik. R-kwadraatwaardes van lineêre regressiemodelle het getoon dat die RPAS-data nie die fenotipiese data in die veld kon voorspel nie, behalwe vir opbrengs. Strooi-plot matrikse het getoon dat daar geen korrelasie was tussen die gegewens wat deur die tradisionele instrumente en die RPAS vasgelê is nie. Die veld- en na-oesdata het aangedui dat 'n naaste buuranalise (NNA) die beste opsie was tydens hierdie studie, aangesien daar veldneigings was en die data is gebruik in die seleksieproses. Toekomstige studies moet die toevoeging van molekulêre merkers insluit wat ooreenstem met verhoogde opbrengsverwante eienskappe om meer ingeligte besluite tydens seleksie te neem. Die ontwikkeling van kameras en sagteware vir afstandswaarneming sal die instrument beslis bevoordeel. Bykomende vegetatiewe indekse kan ondersoek word en die model kan mettertyd verbeter word deur data by te voeg.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/108223
This item appears in the following collections: